Identity Economics is Bad Economics

Identity economics is bad economics. The excellent Nick Rowe puts it well:

Animals can be divided into Carnivores and Non-Carnivores: A = C + NC. Therefore, if we add some wolves to an island of sheep, the number of animals on that island will increase.

It’s easy to see why that argument might not be right. Wolves kill sheep. But if you didn’t know that fact about wolves and sheep, the argument looks very appealing. But the equation A = C + NC tells us absolutely nothing about the world; it’s an accounting identity that is true by definition. The only thing it tells you is how I have chosen to divide up the world into parts. And I can choose an infinite number of different ways to divide the world up into parts.

Here are two more examples:

1. Y = C + I + G + X – M. Therefore an increase in Government spending will increase GDP.

2. Y = C + S + T. Therefore an increase in Taxes will increase GDP.

My guess is that you are much more uncomfortable with the second of those two examples than the first. You have probably seen the first argument before, but have probably not seen the second. But they are both equally correct accounting identities and are both equally rubbish arguments.

As Nick notes, the identities we write down frame our thinking. Write an identity in a different way to check whether the frame really fits.


Add the shoe industry and remove G. Then find solutions along the Shoe axis to set interest charges.

Shoes work great in an accounting identity because we all have a predictable relationship with shoes. We would need a shoe futures market, but once the Fed announces the Shoe standard for an inflation index, then a futures market should develop.

And if there's hyper-inflation, do we call it a 'shoe leather cost'?

Bonus trivia: I solved the first time I heard it the famous: wolf + sheep + grains river crossing logic puzzle. Could you?

Sheep across, empty back.
Wolf across, sheep back.
Grain across, empty back.
Sheep across and you’re done.

Not exactly the hardest puzzle, takes longer to set up than to solve.

There's a fly in your ointment. You do realize that dog food is largely grain-based carbohydrates right? So what's to stop the wolf from eating the grain? Back to the drawing board for you...

Dogs have, through human contact, have evolved carbohydrate tolerance. Wolves have not. They might eat the grain, but they'd get the DIABEETUS.


DIABEETUS. Jeebus, that's the most interesting diseese I never heard of.

This is why economics will never gain the respectability of science. It hates even the most simple of equations. Accounting identities are apparently rubbish. Addition is to be abhorred, subtraction is to be slandered. To parrot what the thinktanks expected you to say is apparently the one true science.

'Wolves kill sheep'

Among other things, but seemingly, we are leaving mice and other animals out of this simplistic - oops, 'simplified' example.

Further, wolves tend to kill unfit sheep, and keep the sheep from overpopulating the island without check - admittedly, this uncontrolled breeding without carnivores leads to more animals. Well, until the population crash.

'it’s an accounting identity that is true by definition'

Using a simplistic - oops, 'simplified' definition. Some might even suggest tautology is the actual word being sought,

'The only thing it tells you is how I have chosen to divide up the world into parts.'

While completely ignoring any complexity in making a simplistic - oops, 'simplified' point.

And you really missed the point of the most. All of your criticisms are correct and that is what he was mocking.

One cannot formulate fiscal or monetary policy based on the Keynesian model without having a sophisticated understanding of how each argument in the identity interacts with the others. And the relationships might be so complex that no stable policy strategy can be formulated. The policy responses are endogenous to the system. Economic actors RESPOND to policies in self interested ways. They avoid and evade taxes, move, work off the books, take payment in unreported cash, substitute leisure, move production offshore, make riskier loans, etc. This can make the planner's solution worse than leaving the economy alone, and possibly make it disastrous.

But the author made zero effort to explain the impact on "complements" and "substitution".

Instead the author did the classic free lunch economics tricks.

For example, conservatives, many economists argue for ending Social Security.

Ok, given estimates are 80% getting benefits depend on SS for at least half their income, they never go on to explain how wonderful it will be for GDP when tens of millions become homeless and hungry unless they move in with their children and grandchildren like when I was born for large numbers of people (1947).

With older workers being shunned by employers for being older than the hiring manager, with past pay twice that if the hiring manager, its wishful thinking they will easily replace benefits with paid jobs. Maybe the 75 year old with a 15 year old car will become a top earning Uber "driver". Odd given Uber claims vision is not enough to nagivate requiring Lidar, so that means older people with glaucoma must be ideal Uber drivers.

Of course, many argue that SS is failing because the number paying taxes is falling relative to the number getting benefits, so better to use individual funds of stocks and bonds.

So, with more people selling stocks and bond starting this year in a hypothetical universe where FICA was used to buy stocks and bonds, the stock market woukd soar because prices go up when more people are selling than buying goods that have no marginal costs. Ie, marginal costs do not cut supply of stocks and bonds sellers are truing to sell, unlike the marginal cost of planting and harvesting soybeans to sell when buyers have dried up.

A century ago, old people retired on small "farms", ie, they owned land on which they grew food, living in houses they had paid off debt, and as outside work for wages fell, they consumed "for free" because eating eggs, garden foods, often canned or stored in root cellars, the cost is zero as is production, to economists.

Foor economists calling for cutting costs, the idea economy has a GDP of zero as zero labor costs are paid, thus consumers pay zero for consumption goods, being forced to live "off the land", consuming only what they produce, perhaps enriched by barter.

The author is trying to argue economics need not be zero sum, but economies must be zero sum. You can't consume in your accounting something not produced in your accounting.

Keynes simply argued that zero sum did not require production go down when labor is idle, but that by finding a way to pay all idle labor, GDP will keep rising until no labor is idle. And, building capital always requires paying workers, with capital limited only by the willingness of workers to save, eg, sacrifice consumption, to create the funds used to pay workers to build capital, like roads, schools, educated workers, more factories, all of which drive down returns to capital.

Those opposed to Keynes policy want workers not paid, thus cutting consumption, thus driving down GDP. But that is not accepted by voters, so the GOP slashes taxes and increasing military spending, paid for by borrowing from workers globally so US consumers buy more than they produce, but with high profits boosting GDP.

garden foods, often canned or stored in root cellars, the cost is zero

I guess you're right, but you've never canned which consumes vast amounts of energy and ongoing storage costs or retrieved from a root cellar.

I guess you're right

garden foods, often canned or stored in root cellars, the cost is zero as is production,

Are those two arguments equivalent? For one thing, isn't the second identity assuming a closed economy while the first has exports and imports?

You're missing the forest for the trees. The arguments are analogous. They need not be identical or even close.

The point is that the terms of the model are interactive in a way that defeats social planning.

But I am not sure anyone really makes the arguments he says. There are those who use the identities to show the breakdown of flows in the economy, but that's different.

Are those two arguments actually equivalent? Isn't the second one assuming a closed economy while the first one accounts for exports and imports?

Opps sorry for the double post, I didn't think my first post went through.

But what if the wolves identify as sheep? These "trans-sheep" aren't true sheep yes, and will still eat sheep, although we wouldn't be allowed to talk about it on twitter lest we get banned.

I agree with clockwork_prior...vast oversimplification. Stop hating trans-sheep, you bigot.

'although we wouldn't be allowed to talk about it on twitter lest we get banned'

I doubt twitter bans anyone for discussing such issues, but since I do not use twitter - and feel that anyone using twitter deserves everything that comes from using twitter - who knows? Links would be nice, but this is the MR comment section, where links are often apparently seen as just another sign of trolling.

But it is amusing how many people complain about a private company providing a free service deciding to act like a private company and make any decisions it wishes concerning that freely offered service.

The Internet is not google, facebook, amazon, and twitter, regardless of how many people seem to believe that. Oddly enough, a group that mainly seems to use google, facebook, amazon, and twitter to a major extent.

However, one has to be honest in pointing out you actually seemed to miss the identity equation - Carnivores and Non-Carnivores: A = C + NC. Basically, the sheep can call themselves wolves too, though they will still eat grass. The names do not change one's identity as C or NC.

Seems like someone wasn't paying attention to the importance of the C / NC distinction, where it is the diet that matters, not whether one is called a sheep or a wolf.

Expert Psychologist Specializing in Gender Dysphoria Suspended From Twitter For Publishing... Science

Apparently though - if you've got supporters (even if they oppose your view) with followers in the 5,000+ class - you can, of course, be reinstated. It appears that you missed the entire point of my facetious post. My point was to call attention to the importance of the mathematical distinction by specifically highlighting metaphorically arguments currently being made that are not mathematical nor scientific.

It is the diet (innate natural properties) that matters, not whether "one is called a sheep or a wolf".

Tell that to Twitter.

'Tell that to Twitter.'

Why? I don't care what some private company does when it comes to deciding who gets to use its free service and who doesn't.

And I have absolutely no idea why I should care about a private company acting the same as a private homeowner, who is free to keep out anyone the homeowner pleases, for any reason at all.

However, if you really care about ensuring no 'censorship' (already a misleading word, but not exactly improper) of ideas occurs through the actions of a private company free (and legally entitled) to take those actions, then I suggest exploring Mastodon -

Another example of a disputed identity is Harrod Domar equation I guess.

Gee, I wonder if that's why every econ textbook also introduces the idea of ceteris paribus? Please cite one economist who asserts that increasing G also increases Y, but who doesn't include the ceteris paribus caveat.

Maybe you think the caveat is not well supported. But in that case, you should provide some evidence, not an argument by analogy (i.e. in this case, Rowe would need evidence that C, I, G, X, and/or M exist in a sheep/wold type relationship).

But even evidence of that relationship wouldn't be enough. Rowe's analogy plays on our mood affiliations about predators and taxes. But there's no reason to suspect, without evidence, that adding wolves would be bad for the animal population. If, for example, wolf reproduction was 2 wolves born for every 1 sheep eaten, then yes, an increase in the wolf population would increase the total animal population. If $1 in extra taxes reduced saving and consumption by less than $1 on net, perhaps because it was spent on R&D projects that the private sector would not undertake, then yes, increasing taxes would increase GDP.

Rowe doesn't get to trot out an analogy about wolves in order to assert 100% crowding out. That's an empirical question. Talk about rubbish arguments.

If you think Nick is defending 100% crowding out you have fallen prey to the very error he identified.

I don't think so. It seems to me that the correct interpretation of his wolf/sheep analogy is that if you are arguing for less than 100% crowding out, the burden is on you to prove it. I would, in fact, characterize that as a defense of the 100% crowding out argument.

I'm saying the existence of the accounting identity is sufficient to justify the claim that people arguing for more (or less) than 0% crowding out are the ones who should have the burden of proof. The accounting identity holds (by definition) as long as there is not endogenous relationship between the subcomponents of the identity. The implication of Rowe's argument is that we must assume the existence, direction (positive crowding out) and magnitude (100% crowding out) of that endogenous relationship, even without any evidence whatsoever. And his justification for that assumption is no better than "wolves eat sheep."

You can't rubbish an argument by doing nothing more than swapping the burdens.

Yes, you can rubbish it. First, the simple Keynesian model controls for all else. The fiscal multiplier is positive and greater than 1.

Second, he is showing a rhetorical equivalence between two fallacious arguments. Details don't matter.

Third, the burden remains on Keynesians to demonstrate the efficacy of fiscal policy. All the hand waving or mathematical treatises can be wiped out by a single counterexample which he has provided.

The flaw in the Keynesian (planner's) solution comes from a fatally flawed presentation of the problem. The wolf and sheep analogy distills this to simple terms. Adding complexity does not rescue it.

#1 is the Expenditure approach to GDP calculation where GDP = Y, Consumption = C, investment = I, gov. spending = G, net exports = X-M

Prof. Tabarrok, what is #2? T is taxes, but what is S? I assume it's savings rate. Increasing Treduces C and S ( savings rate). However, provided there's a lot of absolute existing savings (wealth, real estate/property, cash, equities), a tax increase on the savings rate increases GDP (during the calculated period) and reduces the total amount of savings.

The system works provided the increase in taxes do not deplete the absolute amount of savings.

I don't care about the morals or politics. I just see that Mr. Rowe might be confusing the savings rate with total savings/wealth.

I'll repeat here the post I left at WCI: I think what's being missed here is that an equation expressing a relationship among variables is not very useful (and certainly cannot be counted on to reflect reality) unless all of the variables can be precisely defined and accurately measured. That's where most non-physical science equations fall apart. Just look at all the distorted derivations (and explanations) of sectoral balances that exist.

Its just another Worthwhile Canadian Initiative.

I like Mike Rowe.

A less simplistic proposal with more variables: If you add a windmill farm (Climate Change!) to an island of puffins, the number of puffins on that island will increase. And, you will have a puffin cemetery.

Do-Gooders Doing Damage. Save the puffins from people trying to save the planet.

Both arguments of course are false based on these identities but I still don't think that using accounting identities is wrong. It's still useful to realize that income goes into different "buckets". It might be primitive, but I think many people don't even realize this actually and just think it vanishes.

If you're comfortable with #1, especially if that spending is on the military, then you might be a conservative. If you're comfortable with #2, then you could be a progressive. If both put you in a bad mood and you're a well armed sheep, then you're potentially a libertarian.

Sometimes I feel more like a sheep that wishes he was well armed.

Clicked on this article hoping to see a stance taken against the likes of Wu, Wolfers, and Card.

MMT advocates don't use identities to advance arguments the way Nick implies. They use identities to enforce basic stock/flow consistency, and to point out when others are not doing so.

Wait... I thought this post was going to be about identitarian economics.

Now I'm sitting here disappointed with my untouched bucket of popcorn. Ah well, there's always the next post for shitshow drama in the comments section.

That tax example is wrong. If you increase tax revenue and do not decrease consumption or savings, of course GDP went up!

The identity works, the linked article is drivel.

Look people. This is simple.

These identities are not "equations" in the usual sense. The variables on the right are not independent of each other. They all depend on other things, and have complicated interactions. No one of them changes independently.

So it's nonsense to say that, for example, an increase in T will increase GDP. Any increase in T will affect C and S, which are functions of T (and other variables).

Someone tell Peter Navarro.

Yes: the number of carnivores is a function of the number of non-carnivores, and the number of non-carnivores is a function of the number of carnivores:

A(t+1) = C(NC(t)) + NC(C(t))

True but the identity forces us to think about the dynamics. If I add wolves to the island initially there will be more animals. If later on the total animals are the same then either the sheep had to go down, or the wolves went down, or both went down to offset my increase.

If taxes go up but savings or consumption is reduced to pay the taxes, then Y will not increase. But then that generates a hypothesis. If tomorrow a tax increase is passed it is telling us we should be looking at indicators of savings and consumption. If those indicates do not veer downwards, then the equation would force the counter-intuitive result of Y increasing.

Comments for this post are closed